基于机器学习方法的小微企业信用风险评估问题研究——关于嘉兴市小微企业2013-2017年金融信用信息的实证分析

On Credit Risk Assessment of Small and Micro Enterprises Based on the Method of Machine Learning——An Empirical Study of the Financial Credit Information of the Small and Micro Enterprises in Jiaxing from 2013 to 2017

以嘉兴市2013-2017年发生信贷业务的小微企业为研究对象,对2646家小微企业2013-2017年的金融信用信息进行实证分析,构建了5类小微企业信用风险评估模型,结果显示,随机森林模型更适用于嘉兴市小微企业信用风险评估,根据该方法将小微企业风险分为“高档”“中档”“低档”3类,提出了政策建议。

Taking the small and micro enterprises with credit business in Jiaxing from 2013 to 2017 as the research objects,we carry out an empirical analysis of the financial credit information of 2646 small and micro enterprises among them.Five kinds of credit risk assessment models are established and the results show that the random forest model is more suitable for the assessment of the small and micro enterprises in Jiaxing.According to this method,small enterprises are classified into three categories,namely,“the high-grade”,“the mid-grade”and“the low-grade”,and policy suggestions are put forward.